Squeeze3D: Your 3D Generation Model is Secretly an Extreme Neural Compressor

• URL: http://arxiv.org/abs/2506.07932v1
• Date: Mon, 09 Jun 2025 16:52:10 GMT
• Title: Squeeze3D: Your 3D Generation Model is Secretly an Extreme Neural Compressor
• Authors: Rishit Dagli, Yushi Guan, Sankeerth Durvasula, Mohammadreza Mofayezi, Nandita Vijaykumar,
• Abstract summary: We propose Squeeze3D, a novel framework that leverages implicit prior knowledge learnt by existing pre-trained 3D generative models to compress 3D data at extremely high compression ratios.<n>Our approach bridges the latent spaces between a pre-trained encoder and a pre-trained generation model through trainable mapping networks.<n>Our experiments demonstrate that Squeeze3D achieves compression ratios of up to 2187x for textured meshes, 55x for point clouds, and 619x for radiance fields while maintaining visual quality comparable to many existing methods.
• Score: 5.06976177768381
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We propose Squeeze3D, a novel framework that leverages implicit prior knowledge learnt by existing pre-trained 3D generative models to compress 3D data at extremely high compression ratios. Our approach bridges the latent spaces between a pre-trained encoder and a pre-trained generation model through trainable mapping networks. Any 3D model represented as a mesh, point cloud, or a radiance field is first encoded by the pre-trained encoder and then transformed (i.e. compressed) into a highly compact latent code. This latent code can effectively be used as an extremely compressed representation of the mesh or point cloud. A mapping network transforms the compressed latent code into the latent space of a powerful generative model, which is then conditioned to recreate the original 3D model (i.e. decompression). Squeeze3D is trained entirely on generated synthetic data and does not require any 3D datasets. The Squeeze3D architecture can be flexibly used with existing pre-trained 3D encoders and existing generative models. It can flexibly support different formats, including meshes, point clouds, and radiance fields. Our experiments demonstrate that Squeeze3D achieves compression ratios of up to 2187x for textured meshes, 55x for point clouds, and 619x for radiance fields while maintaining visual quality comparable to many existing methods. Squeeze3D only incurs a small compression and decompression latency since it does not involve training object-specific networks to compress an object.

Related papers

• Harnessing Text-to-Image Diffusion Models for Point Cloud Self-Supervised Learning [61.04787144322498]
  Diffusion-based models, widely used in text-to-image generation, have proven effective in 2D representation learning.<n>We propose PointSD, a framework that leverages the SD model for 3D self-supervised learning.
  arXiv  Detail & Related papers  (2025-07-12T01:20:07Z)
• Mesh Compression with Quantized Neural Displacement Fields [31.316999947745614]
  Implicit neural representations (INRs) have been successfully used to compress a variety of 3D surface representations.<n>This work presents a simple yet effective method that extends the usage of INRs to compress 3D triangle meshes.<n>We show that our method is capable of preserving intricate geometric textures and demonstrates state-of-the-art performance for compression ratios ranging from 4x to 380x.
  arXiv  Detail & Related papers  (2025-03-28T13:35:32Z)
• DIRECT-3D: Learning Direct Text-to-3D Generation on Massive Noisy 3D Data [50.164670363633704]
  We present DIRECT-3D, a diffusion-based 3D generative model for creating high-quality 3D assets from text prompts. Our model is directly trained on extensive noisy and unaligned in-the-wild' 3D assets. We achieve state-of-the-art performance in both single-class generation and text-to-3D generation.
  arXiv  Detail & Related papers  (2024-06-06T17:58:15Z)
• Compress3D: a Compressed Latent Space for 3D Generation from a Single Image [27.53099431097921]
  Triplane autoencoder encodes 3D models into a compact triplane latent space to compress both the 3D geometry and texture information. We introduce a 3D-aware cross-attention mechanism, which utilizes low-resolution latent representations to query features from a high-resolution 3D feature volume. Our approach enables the generation of high-quality 3D assets in merely 7 seconds on a single A100 GPU.
  arXiv  Detail & Related papers  (2024-03-20T11:51:04Z)
• LN3Diff: Scalable Latent Neural Fields Diffusion for Speedy 3D Generation [73.36690511083894]
  This paper introduces a novel framework called LN3Diff to address a unified 3D diffusion pipeline. Our approach harnesses a 3D-aware architecture and variational autoencoder to encode the input image into a structured, compact, and 3D latent space. It achieves state-of-the-art performance on ShapeNet for 3D generation and demonstrates superior performance in monocular 3D reconstruction and conditional 3D generation.
  arXiv  Detail & Related papers  (2024-03-18T17:54:34Z)
• DatasetNeRF: Efficient 3D-aware Data Factory with Generative Radiance Fields [68.94868475824575]
  This paper introduces a novel approach capable of generating infinite, high-quality 3D-consistent 2D annotations alongside 3D point cloud segmentations. We leverage the strong semantic prior within a 3D generative model to train a semantic decoder. Once trained, the decoder efficiently generalizes across the latent space, enabling the generation of infinite data.
  arXiv  Detail & Related papers  (2023-11-18T21:58:28Z)
• PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm [111.16358607889609]
  We introduce a novel universal 3D pre-training framework designed to facilitate the acquisition of efficient 3D representation.<n>For the first time, PonderV2 achieves state-of-the-art performance on 11 indoor and outdoor benchmarks, implying its effectiveness.
  arXiv  Detail & Related papers  (2023-10-12T17:59:57Z)
• GaussianDreamer: Fast Generation from Text to 3D Gaussians by Bridging 2D and 3D Diffusion Models [102.22388340738536]
  2D and 3D diffusion models can generate decent 3D objects based on prompts. 3D diffusion models have good 3D consistency, but their quality and generalization are limited as trainable 3D data is expensive and hard to obtain. This paper attempts to bridge the power from the two types of diffusion models via the recent explicit and efficient 3D Gaussian splatting representation.
  arXiv  Detail & Related papers  (2023-10-12T17:22:24Z)
• GET3D: A Generative Model of High Quality 3D Textured Shapes Learned from Images [72.15855070133425]
  We introduce GET3D, a Generative model that directly generates Explicit Textured 3D meshes with complex topology, rich geometric details, and high-fidelity textures. GET3D is able to generate high-quality 3D textured meshes, ranging from cars, chairs, animals, motorbikes and human characters to buildings.
  arXiv  Detail & Related papers  (2022-09-22T17:16:19Z)
• Implicit Autoencoder for Point-Cloud Self-Supervised Representation Learning [39.521374237630766]
  The most popular and accessible 3D representation, i.e., point clouds, involves discrete samples of the underlying continuous 3D surface. This discretization process introduces sampling variations on the 3D shape, making it challenging to develop transferable knowledge of the true 3D geometry. In the standard autoencoding paradigm, the encoder is compelled to encode not only the 3D geometry but also information on the specific discrete sampling of the 3D shape into the latent code. This is because the point cloud reconstructed by the decoder is considered unacceptable unless there is a perfect mapping between the original and the reconstructed
  arXiv  Detail & Related papers  (2022-01-03T18:05:52Z)
• Deep Hybrid Self-Prior for Full 3D Mesh Generation [57.78562932397173]
  We propose to exploit a novel hybrid 2D-3D self-prior in deep neural networks to significantly improve the geometry quality. In particular, we first generate an initial mesh using a 3D convolutional neural network with 3D self-prior, and then encode both 3D information and color information in the 2D UV atlas. Our method recovers the 3D textured mesh model of high quality from sparse input, and outperforms the state-of-the-art methods in terms of both the geometry and texture quality.
  arXiv  Detail & Related papers  (2021-08-18T07:44:21Z)
• PvDeConv: Point-Voxel Deconvolution for Autoencoding CAD Construction in 3D [23.87757211847093]
  We learn to synthesize high-resolution point clouds of 10k points that densely describe the underlying geometry of Computer Aided Design (CAD) models. We introduce a new dedicated dataset, the CC3D, containing 50k+ pairs of CAD models and their corresponding 3D meshes. This dataset is used to learn a convolutional autoencoder for point clouds sampled from the pairs of 3D scans - CAD models.
  arXiv  Detail & Related papers  (2021-01-12T14:14:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.